Is there any stochastic formulas estimating the probability of where the stock will go over x amount of time? ie: if a stock has been going down over the past week, what is the probability it continues to go down tomorrow with no additional information.
Not only does such a formula exist, I would bet a million of them exist. Which one is correct, that's the question. The second question is, does it matter which one is correct or can you trade without it. Consider this scenario: what if your formula said that a certain stock would go down tomorrow, from today's close, so you shorted it at the market close. But, let's say the stock roars up 10% during the first hour of trading and stays there all day until just before the close, then declines 11% in the last minute of trading. Would you have held on to it the entire day to get that 1% return? Huge difference between forecasting and trading. Maybe you know that already, at least in theory, but if you don't consider that fact at all times, success will be very difficult to achieve. Forecasting is far less important than identifying a set-up and how you will manage the trade from that set-up. Not trying to throw you off, just trying to steer you toward what's important.
I don't subscribe to real-time stock data for charting as I trade futures. But attached is the picture for JPM as of yesterday's close. If it goes below the purple 31.34 line in dollars it is a high probability short. The chart also shows euro pricing.
Yeah, I just thinking about projecting about what I'd do now, not based on the future. Then re-evaluate what I'd do with the trends change. What I gave was an example, it doesn't have to be a week and could be any finite length of time. Thanks.
The solid blue line moving lazily through the chart is the 150 day sma of the close, which a lot of mindless idiots think is significant. The fourth pane from the bottom is the dollar/euro exchange rate, which a different bunch of idiots think drives the market. The rest is my own empirical stuff, unfortunately only believed by one idiot.
Right, and my scenario could also be a week or any finite length of time. There are people who have been insisting that the US market is a short since March of 2009. Whether they are actual traders (or still traders) is a different story. They have lots of forecasts to back them up, yet here the market is, up 100% since then. That's an extreme example, but there were plenty of people who finally woke up and ditched their forecast after 6 months of rallying from the March 2009 low, too. Because of a forecast, they missed 6 of the best months in US equity market history. And these aren't stupid people either, and I bet a lot of them, if not all, figured they would "re-evaluate" when "the trends change". If you want to trade and make money from trading, what you need to do is find a set-up that has positive expectancy and figure out how to manage the trades that result from that set-up. It's actually a lot harder than forecasting, believe it or not. That's why forecasters get paid a fraction of what a good trader gets paid. Forecasters may make $1 million/year if they are at the absolute pinnacle of the profession. Traders can bring home $100 million or more at that same pinnacle of their profession. It's like the saying in golf, "Drive for show, putt for dough". Forecasting is driving, trade identification and management is putting. You should check out the ACD Method thread for a way of using set-up management. It's as good as anything out there in the public domain. I wish someone had told me this when I was new to trading. Or, maybe I'm misunderstanding your intention when you talk about "predicting" what a stock will do.